An IEA-ETS Research Institute Journal
From: The measure of socio-economic status in PISA: a review and some suggested improvements
Original (ESCS) | Replica (ESCS1) | Alternative 1 (ESCS2) | |
---|---|---|---|
“Parental education” | PARED | PARED | PARED |
“Parental occupation” | HISEI | HISEI | HISEI |
“Income/wealth” | HOMEPOS | HOMEPOS1 | HOMEPOS1 |
Imputation of missing values | Regression on non-missing components + random normal noise | ||
Weighting scheme | PCA | PCA | Equal weights |
Non-missing observations | 466,458 | 466,441 | 466,441 |
missing observations | 10,394 | 10,411 | 10,411 |
OECD mean | − 0.04 | 0.00 | 0.00 |
OECD st.dev. | 1.00 | 1.00 | 1.00 |
Range (min, max) | − 7.3; 4.2 | − 6.8; 4.0 | − 6.9; 4.0 |
Correlation with original | 1.00 | 0.99 | 0.99 |
Country means | |||
Pearson correlation with original | 1.00 | 0.99 | 0.99 |
Spearman correlation with original | 1.00 | 0.99 | 0.99 |
Mean absolute deviation with original | 0.00 | 0.04 | 0.04 |
Max positive deviation and country | 0.00 | 0.34 (Lebanon) | 0.33 (Lebanon) |
Min negative deviation and country | 0.00 | − 0.01 (Qatar) | − 0.01 (Qatar) |
Relationship with science performance (R2) | |||
OECD average | 0.129 | 0.132 | 0.132 |
Pearson correlation with original | 1.00 | 1.00 | 1.00 |
Spearman correlation with original | 1.00 | 1.00 | 1.00 |
Mean absolute deviation with original | 0.00 | 0.00 | 0.00 |
Max positive deviation | 0.00 | 0.01 (Slovak Republic) | 0.01 (Slovak Republic) |
Min negative deviation | 0.00 | − 0.01 (Belgium) | − 0.01 (Belgium) |
Relationship with science performance (beta) | |||
OECD average | 38.4 | 38.4 | 38.4 |
Pearson correlation with original | 1.00 | 1.00 | 1.00 |
Spearman correlation with original | 1.00 | 1.00 | 1.00 |
Mean absolute deviation with original | 0.00 | 0.43 | 0.43 |
Max positive deviation | 0.00 | 1.9 (Lebanon) | 2.1 (Lebanon) |
Min negative deviation | 0.00 | − 2.0 (Belgium) | − 1.9 (Belgium) |